The resonant switching topology is, to date, one of the
most efficient solutions for designing switch mode power supplies.


"Although in existence for many years, only recently has the LLC and LCC resonant converter (in particular in its half-bridge implementation) gained in the popularity it certainly deserves. In many applications, such as flat panel TVs, PCs and notebooks, LED drivers, where the efficiency, power density and cost requirements are getting increasingly stringent, the resonant half-bridge with its many benefits and very few drawbacks is an excellent solution. Also in applications with large ouput voltage ranges, the LCC topology in particular is the most suitable."


The introduction of new regulations, both voluntary and mandatory, has brought about a revaluation of the LLC and LCC resonant topology.

In LED lighting (to name just one of the infinite applications) the need for more and more efficient power supply systems is pushing power designers in this direction.

All the most important manufacturers of active components on the SMPS controller market have included efficient chips in their product catalogues. Also some new families of Mosfets with specific characteristics for resonant converters have been developed.

With an effectively contained degree of circuit complexity, they allow the realization of power supplies with 93-96% efficiency (which can be further improved using synchronous rectifiers instead of rectification diodes) and reduce EMI/EMC issues compared to other topologies thanks to "Zero Voltage Switching" and to the substantially sinusoidal high frequency currents.

"Generally speaking, resonant converters are switching converters that include a tank circuit actively participating in determining input-to-output power flow."

The operating principle is based on the characteristic gain curve of the resonant tank", which allows to change the gain through a moderate variation of the switching frequency, thus resulting in an effective regulation of output voltage or current in relation to load and input voltage changes.

The resonant "tank" is a set of two inductive elements and one capacitor (LLC) or two inductive elements and two capacitors (LCC).

Using a conventional transformer, the two inductive elements are the magnetizing inductance of the transformer and a discrete resonant inductor. On the other hand, the “integrated transformer” exploits part of the leakage inductance, eliminating the need for a discrete resonant inductor.

The use of an integrated transformer is therefore much more convenient in terms of cost, size and energy efficiency.The specific features of the integrated transformer are described below.

To give an idea of the advantages, a well-designed 180 W integrated transformer can measure less than 28x29x23 mm, with costs that are obviously more competitive compared to solutions that use a discrete inductance.

It is evident that the only possible reason leading to the use of a convetional transformer is the difficulty of designing a coherent and optimized tank.

In most electronic equipment from the largest manufacturers the inductive components are well structured. Usually in these cases a time-consuming optimization is performed through FEM tools. Despite potentially having access to the same technologies, many other manufacturers make use of very poor magnetic components. The level of optimization is not the same because of a lack of highly specific skills, efficient design tools or big budgets. Sometime all of these.


- typical range of efficiency for the simplest circuitry 94-96%, with possible improvements through synchronous rectification and other small adjustments;

- using correctly sized magnetic components, the design is rapid and much simplified;

- the current waveform at high frequencies is basically sinusoidal, with significant reductions in harmonics compared to other topologies;

- MOSFET commutation ON "ZVS" (Zero Voltage Switching) with associated elimination of commutation loss, reduction/elimination of dissipators and reduction of stress and EMI, which often cause the most hostile design problems;

- the possibility to reduce consumption with low/zero load by using the functions "burst mode" and "PFC stop" implemented on many controllers;

- the possibility of optimal sizing for continuous and temporary power, including some significant improvements on conventional solutions (e.g. an optimised transformer with volume similar to a classic EF25 can easily supply up to 500Wpk and more);

- compared to other topologies, LLC and LCC power supplies have smaller dimensions with a notable reduction in EMI/EMC issues;

- typical cost savings on dissipators, EMC filters, smaller transformers-PCB-enclosures.



- the cost and performance of a resonant power supply largely depends on the the tank and the inductive components. Even a good level of skill may not be enough for an optimal design;

- the resonant controller is more costly than the flyback counterpart;

- two MOSFETs (half bridge) are required, rather than the single one required for the Flyback topology

(however, even disregarding the efficiency improvement, the related cost increase tends to cancel itself - or even turn to a saving - thanks to the cost reduction on other components).


Lots of documents are available to extend the topic; let's take a look at some examples.

Here are some quick design considerations from  Fairchild Semiconductor ®  and  Texas Instruments ®.

Here are some useful and more comprehensive application notes from ST Microelectronics ®Fairchild Semiconductor ® and Infineon ®.



The so-called "integrated resonant transformers", make use of leakage inductance (which normally represents an undesirable parasitic effect) instead of a discrete inductor, integrating two of the three resonant tank elements in just one inductive component.

In addition to convenience in terms of cost and dimensions, it must be highlighted that the magnetic flux of the leakage inductance fundamentally travels through free air, thus eliminating every problem linked to the core saturation. For a discrete inductor it is not the case.

In order to achieve good results, the design structure and details must be skillfully managed so as to obtain the required leakage inductance, in relation to all the other design parameters, under conditions of minimal power loss.

In many cases, empirical experience and generic methods of calculation can lead to acceptable approximations; but not in those applications where high efficiency is strictly required.

In these cases, a few more lost Watts - or sometimes a fraction of a Watt - can significantly affect the power supply's overall efficiency. It can easily compromise the careful choices made during the design of the converter.

Optimal efficiency for inductive components can only be achieved by surpassing a number of simplified design methodologies, such as the equal division of the power loss target between the core and the copper.

Literature and experience teach us that the best efficiency point can be identified via the "ad hoc" definition of power loss depending on several trade-offs.

In the specific case of integrated transformers, there are a number of restrictions requiring close cooperation with the manufacturer of inductive components.

The definition of the best parameters for a resonant tank cannot be made without considering the restrictions linked to the structural elements of each transformer - first and foremost the curve showing the relationship between inductance and leakage inductance.

Without this dialogue, you will - at best - be forced to work with an inappropriate inductance value, which can result in a bad design.

The most critical issue in the design of integrated transformers is the realistic calculation of winding losses, without which any design optimization becomes unfeasible. With this calculation, the Eddy Current loss resulting from the "proximity effect" should be considered as well as the "skin effect".

A lack of consideration of these issues, or a lack of know-how in magnetic component design, often leads to designing low-efficiency transformers (and power supplies) from an economical, energy and dimensional standpoint, with an undesirable waste of resources.


The following comparative test reports show how it's possible to increase efficiency thanks to an optimized integrated transformer.

Take a look at some interesting examples:

Other examples here:


For more info about the integrated resonant transformers click on the following links:




The active PFC stage on the input of a SMPS is often mandatory.

There are also some critical design considerations about this component, especially for some of the most used types.

The most popular type of PFC adopted for power levels up to 200-300 W is the "Transition Mode" (also named "Critical Mode" or "Boundary mode"), where the usual core loss calculation methods cannot be used due to the complexity of the current waveform.

In fact, even at a constant load, the current has an almost triangular wave shape, but with continuously variable frequency and amplitude depending on the instantaneous input voltage value [|sen (VinRMS)|].

On the other hand, with “continous mode” (CCM) PFC flux sub-loop and variable DC bias occurs, introducing other critical aspects on the power loss calculation.

It must be considered that the power loss curves published by the manufacturers of magnetic cores refer to sinusoidal waveforms as well as specific frequencies and temperatures and therefore are not directly applicable. In this case advanced calculation methods are needed.

Additionally, the previously mentioned problems about the calculation of power loss in the windings also exist for the PFC inductor.

At this point it is clear that an optimal design requires the access to specific resources and tools, often not available in electronic design teams.


More info about PFC inductors here:


Everything about PCB transformers and inductors on

itacoil website

ITACOIL srl - VAT IT03071100964

ITACOIL and registred trade marks.